{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/boosting-unknown-number-speaker-separation","title":"Boosting Unknown-number Speaker Separation with Transformer Decoder-based Attractor","arxiv_id":"2401.12473","date":"2024-01-23","proceeding":null,"authors":["Younglo Lee","Shukjae Choi","Byeong-Yeol Kim","Zhong-Qiu Wang","Shinji Watanabe"],"abstract":"We propose a novel speech separation model designed to separate mixtures with an unknown number of speakers. The proposed model stacks 1) a dual-path processing block that can model spectro-temporal patterns, 2) a transformer decoder-based attractor (TDA) calculation module that can deal with an unknown number of speakers, and 3) triple-path processing blocks that can model inter-speaker relations. Given a fixed, small set of learned speaker queries and the mixture embedding produced by the dual-path blocks, TDA infers the relations of these queries and generates an attractor vector for each speaker. The estimated attractors are then combined with the mixture embedding by feature-wise linear modulation conditioning, creating a speaker dimension. The mixture embedding, conditioned with speaker information produced by TDA, is fed to the final triple-path blocks, which augment the dual-path blocks with an additional pathway dedicated to inter-speaker processing. The proposed approach outperforms the previous best reported in the literature, achieving 24.0 and 23.7 dB SI-SDR improvement (SI-SDRi) on WSJ0-2 and 3mix respectively, with a single model trained to separate 2- and 3-speaker mixtures. The proposed model also exhibits strong performance and generalizability at counting sources and separating mixtures with up to 5 speakers.","url_abs":"https://arxiv.org/abs/2401.12473v1","url_pdf":"https://arxiv.org/pdf/2401.12473v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"speaker-separation","task_name":"Speaker Separation"},{"task_slug":"speech-separation","task_name":"Speech Separation"}],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/speech-separation-on-wsj0-2mix","task":"Speech Separation","dataset":"WSJ0-2mix","model":"SepTDA (L=12)","rank_in_archive_order":6,"of":40,"metrics":{"SI-SDRi":"24.0"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-wsj0-3mix","task":"Speech Separation","dataset":"WSJ0-3mix","model":"SepTDA","rank_in_archive_order":1,"of":9,"metrics":{"SI-SDRi":"23.7"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-wsj0-4mix","task":"Speech Separation","dataset":"WSJ0-4mix","model":"SepTDA","rank_in_archive_order":1,"of":5,"metrics":{"SI-SDRi":"22.0"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-wsj0-5mix","task":"Speech Separation","dataset":"WSJ0-5mix","model":"SepTDA","rank_in_archive_order":1,"of":6,"metrics":{"SI-SDRi":"21.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2401.12473","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}